Into the belly of a Little Red Dot

This paper discusses the Singapore government’s tray return interventions for the island’s hawker centres, and suggests that Design Thinking research techniques can help surface other unaddressed motivations, which could improve the take-up rates of tray returns.

Shirley Lai
Nov 12, 2018 · 6 min read

This paper was written as part of Hyper Island’s part time MA in Digital Management: Design Thinking module.


Singapore’s hawker centres are representations of our multi-racial heritage, food and culture, and are places for Singaporeans to bond over affordable food. In light of the importance of hawker centres to Singapore’s social fabric, the Ministry of the Environment and Water Resources (MEWR) appointed a Hawker Centre 3.0 committee “to review the management of hawker centres and the sustainability of the hawker trade, and work towards the vision of the hawker centres being appealing and vibrant social spaces for the community“. (Hawker Centre 3.0 Committee Report, 2017)

One of the Committee’s focus areas was encouraging tray returns, as this causes slow table turn-overs, and soiled crockery causing hygiene issues is a common occurrence in our hawker centres.

Current interventions have met with limited success: through a conversation with an anonymous spokesperson from MEWR, it was revealed that the Hawker Centre Division is currently measuring the application of RFID tray deposits in hawker centres, and looking to mitigate the unintended consequences (Chua, 2017) (Tang, 2018) of the initiative.

Current behavioural interventions

The Singapore government has implemented, or will be implementing interventions in these ways:

Boosting motivations

  • Appealing to patron’s values with graciousness campaigns, mascots, murals (Khew, 2016)
  • Incentivising with cash-back / free coffee (Khew, 2016)
  • Penalising with RFID-tagged trays (Toh, 2018)

Enhancing ability

  • Reducing barriers by providing more prominent tray return systems and better layout (Hawker Centre 3.0 Committee Report 2017, p40)
  • Reducing barriers by using robot technology (Siau, 2017)

Providing / removing triggers

  • Removing cues by making cleaners inconspicuous (Hawker Centre 3.0 Committee Report 2017, p39)
  • Providing reminders via ambassadors (Sim, 2014), aprons, posters (Khew, 2016) (Boh, 2018)

Some approaches have worked better than others. Measures to nudge Singaporeans gently with visual reminders have been met with resistance, with some patrons setting up a Say NO to Tray Return Singapore Facebook page, or expressing their defiance in personal blogs. Robot tray collectors were ignored by patrons when it moved past their table into other lanes.

Applying Design Thinking

To understand why some interventions have met with limited success, I propose looking into the problem from the user’s perspective. By doing so, we might be able to capture insights that would help mitigate risks and improve current interventions. Given that the Hawker Centre Division is looking into expanding initiatives like RFID tray deposits into 25 other hawker centres (Lim, 2018), new user insights might prove to be cost and time-saving.

The hawker centre is a public space where a large variety of people cross paths daily. The technique chosen for researching physical spaces should be able to surface insights on how a user interacts with their environment, and how the environment influences their actions. I propose the usage of contextual inquiry, a form of field research that allows for in-depth information gathering, with observations happening in a user’s environment (Holtzblatt & Beyer, 1997). The researcher observes and probes a user, with the user in the position of the “expert”, using this technique on a one-to-one basis. Contextual inquiry straddles attitudinal and behavioural research (Rohrer, 2014), allowing us to understand ”what people say” and “what people do”. Daae & Boks (2015) predicts that this combination is suitable for discovering a user’s habits, beliefs, attitudes, intentions, objective and subjective constraints, social and personal norms, and values. Since contextual inquiry is a qualitative method, it is also suited to answer questions of ”why” and “how to fix” a problem (Rohrer, 2014).

Contextual inquiry differs from research techniques highlighted in Hawker Centre 3.0:

  • Surveying is a quantitative method, used to understand “how many” and “how much” questions (Rohrer, 2014). For instance, numbers can inform where the focus of interventions may be, but it might not be able to inform the kind of interventions that should be taken
  • Focus groups sessions provide “a top-of-mind view of what people think about a brand or product concept in a group setting” (Rohrer, 2014), but because discussions are not carried out individually, users may “think they want one thing when they need another”(Nielsen, 1997) or be led into group think, leading to misleading data

In the context of the hawker centre, I surmise that contextual inquiry has the ability to supplement insights that resulted from the above-mentioned techniques, allowing us to discover:

  • Unaddressed factors that will cause a patron to return / not return a tray, and why
  • Social / personal norms they adhere to
  • Attitude towards tray returns and current implementations

Application: Contextual inquiry and user interviews

I conducted contextual inquiry at Old Airport hawker centre in the heartlands, after lunch hours (2pm) on a weekend (Saturday). I observed and spoke to six individuals. I took the chance to briefly interview these users as well to understand their motivations, pains and behaviours. I also made observations in One North’s Timbre+ to see what patrons did when handling RFID tray returns.

All participants were informed of the aim of this project and how long it would take. A handful of people gave consent to audio recordings for transcription purposes. They were also informed that they may withdraw for any reason.

Some insights collected from contextual inquiry:

“I removed the bowls from the tray because it was for two people, we can’t share the tray. If I were just eating alone, I won’t be removing the bowl. ”

“I never return my tray. If they charge me extra for tray deposit, then I have no choice.”

“I’ll feel embarrassed if others clean up after themselves and I don’t follow suit.”

Research insights (Appendix A) indicated that a combination of factors in a patron’s experience is necessary for a successful tray return, instead of a single influence being the catalyst. For example, improving environmental barriers like lane width would increase tray return rates — if cleaners were not seen prominently throughout the eating area. Being gracious is a socially desirable behaviour, but if it interferes with individual concerns like convenience, then the tray return might not happen.

Research also surfaced that there are other influences to behavioural change that were not addressed by current interventions, such as:

Peer influence
Are my fellow patrons returning trays?
Is there someone standing beside my table, waiting to take over my table?

Do I have physical blockers?
Am I with children or the elderly? (see: enhancing ability)

Am I eating alone? Will I even need a tray for the food I’m buying?
Will my tray crowd the table when eating?

Are there birds flying round, picking on uneaten food?
Does this hawker centre look like it’s being taken care of?

What time am I having my meal? Are there cleaners around to assist in quick turnover?

With these insights the Hawker Centre Division can consider moving into affinity mapping, to “bring data from individual customer interviews together so the team can see common pattern and structure without losing individual variation“ (Holtzblatt & Beyer, 1997). The How Might We method (Both, undated) could also be used next to reframe their challenge into new opportunities or improvements. This, together with the Design Sprint framework (Knapp, Zeratsky & Kowitz 2016), would allow the department to test new, risky ideas quickly. (Appendix B)

Areas of improvement for research

  • Research could have been planned with broader observations in mind, starting from when a patron shops for food at different stalls. Current insights were drawn from when a patron has already purchased their food, or when it was the end of a meal.
  • Research could be conducted in more time frames (i.e peak hours, dinner time) , as patrons might face different challenges during these hours.
  • Recruit a variety of users that represents the general population. For this document observations were made in only two hawker centres. Expanding contextual inquiry to a boarder audience might allow us to collect insights that are specific to different areas in Singapore.

Technique application
Contextual inquiry requires one to be non-judgemental, empathetic and friendly. It is also likely that the user will be mindful that she is being observed, thereby leading to unusual actions she might not otherwise do. It is therefore vital that the researcher be discreet, only surfacing to ask questions in areas of interest.


The research technique highlighted here has successfully surfaced unaddressed tray return motivations. If used on a narrower scope (i.e RFID tray deposits), the same technique would likely be able to surface detailed insights the Hawker Centre Division can use, before expanding an idea.

With the government now encouraging co-creation and listening to citizens in policy-making (Chin, 2016) (Teo, 2017), reaching out to the public is more welcomed than ever. I am confident that the user-centred nature of contextual inquiry would be an effective tool in sparking off future ideations and implementations by the Hawker Centre Division.

Shirley Lai

Written by

Research, design & UX writing •

Welcome to a place where words matter. On Medium, smart voices and original ideas take center stage - with no ads in sight. Watch
Follow all the topics you care about, and we’ll deliver the best stories for you to your homepage and inbox. Explore
Get unlimited access to the best stories on Medium — and support writers while you’re at it. Just $5/month. Upgrade